The Clinical Importance of Alloantibody-Mediated Rejection
Article first published online: 28 MAY 2003
American Journal of Transplantation
Volume 3, Issue 6, pages 639–640, June 2003
How to Cite
Halloran, P. F. (2003), The Clinical Importance of Alloantibody-Mediated Rejection. American Journal of Transplantation, 3: 639–640. doi: 10.1034/j.1600-6143.2003.00172.x
- Issue published online: 28 MAY 2003
- Article first published online: 28 MAY 2003
- Received 13 February 2003, revised and accepted for publication 4 April 2003
The minireview by Helmut Feucht and the special article by Paul Terasaki in this issue of American Journal of Transplantation, coupled with the recent addition of alloantibody-mediated kidney rejection to the Banff classification (1), provide an opportunity to reflect on the role of alloantibody against HLA antigens in graft rejection. The emerging message is that donor-specific alloantibody can mediate a number of clinical rejection syndromes, and that C4d deposition is an important tool for recognizing this mechanism.
Peter Gorer (see cover) discovered the mouse major histocompatibility complex (MHC) and demonstrated that allograft rejection consistently evokes antibody production against donor MHC antigens. He observed that mice rejecting allogeneic tumors developed alloantibody against ‘antigen II’ (2,3), later named H-2. Gorer and O'Gorman demonstrated that alloantibodies against MHC are detectable by complement-mediated lysis (4), a property that became the basis of the histocompatibility testing and cross-matching. Gorer believed that alloantibodies played a role in allograft rejection, but acknowledged the major role of lymphocytes, and the variety of effects of alloantibody on experimental allografts, including rapid destruction, slow deterioration, and ‘enhancement’ of graft survival (5). Sometimes alloantibody had little apparent short-term effect because the graft ‘accommodated’ to the presence of alloantibody, a behavior recently associated with the expression of endothelial protective genes.
Dausset and Van Rood first described human sera with alloantibody against what proved to be HLA (6,7). The recognition that antibody against HLA could mediate hyperacute rejection (HAR) [already described for ABO blood group incompatibility (8)] established the destructive potential of preformed alloantibody against HLA (9,10). The development of the microcytotoxicity assay by Paul Terasaki permitted cross-matching and the systematic study of anti-HLA alloantibody (11). Preformed anti-HLA antibody predicted an increased risk of poor initial graft function, and alloantibody appearing during the post- transplant course correlated with increased graft loss, as documented by Terasaki in the accompanying article. But until distinct clinical syndromes were defined, it was arguable that alloantibody was only a marker of T-cell sensitization, and screening for alloantibody after transplantation had little clinical significance.
This changed with the definition of alloantibody-mediated rejection syndromes on the basis of clinical, pathologic, and immunologic findings. Three patterns are emerging: HAR, early alloantibody-mediated rejection; and late alloantibody-mediated rejection. HAR probably reflects the full impact of high levels of preformed antibody rapidly engaging endothelium of the allograft, with the destruction of the vessels and graft necrosis. In contrast, early alloantibody-mediated rejection in kidney presents with graft dysfunction resembling acute tubular necrosis, with inflammatory cells in peritubular capillaries on biopsy, and circulating anti-donor antibody (12,13). These cases typically lack tubulitis, the main feature of T-cell-mediated rejection, and the clinical and pathologic features are relatively nonspecific, overlapping other types of injury such as ischemia. Immunoglobulin is not usually demonstrable on the endothelium in antibody-mediated rejection, for reasons that are not clear. Thus the demonstration that kidneys with antibody-mediated rejection display C4d staining in the peritubular capillaries (14,15) has been a major advance (16). The diagnosis of early antibody-mediated rejection is now securely based on Banff criteria, which include evidence of tissue injury, C4d staining, and demonstration of circulating donor-specific antibody (1).
The features of late alloantibody-mediated kidney rejection and of antibody-mediated injury in other organs are currently being defined. While C4d staining is proving increasingly valuable in studying these cases, as detailed by Feucht in the accompanying article, complete understanding of the impact of alloantibody remains incomplete. For example, C4d staining of peritubular capillaries and circulating donor-specific alloantibody are demonstrable in some cases of renal transplant glomerulopathy but not in others. Does that imply that transplant glomerulopathy is heterogeneous? Many new insights will follow as the roles of alloantibody are re-examined. An indirect benefit of these studies of antibody effects will be to exclude antibody: the absence of C4d staining and of donor- specific antibody helps to make the diagnosis of T-cell-mediated rejection more secure. Nevertheless, the presence of C4d and alloantibody does not exclude T-cell mediated rejection: the two can coexist, although this is not common in our experience.
One surprise in the emergence of antibody-mediated rejection was that arteries were not the usual target of alloantibody. Because early studies (17) associated arterial injury with alloantibody, we have tended to equate arteritis with ‘humoral rejection’. In fact, the inflammatory cells in peritubular capillaries and the pattern of C4d staining indicate that the main target of alloantibody is probably the microcirculation. In contrast, endothelial arteritis in kidney transplantation is often associated with features of T-cell-mediated rejection such as tubulitis (12,13,18). Thus ‘vascular rejection’ is not synonymous with ‘humoral rejection’. At present, the consensus is that either T cells or alloantibody can produce arterial injury, highlighting the need for criteria that distinguish antibody-mediated from T-cell-mediated arterial damage.
Despite new immunosuppressives and improved cross-matching, alloantibody-mediated rejection will increase in importance. Many programs are trying to transplant sensitized patients previously considered untransplantable due to preformed antibody against HLA, and are developing protocols to suppress alloantibody before the transplant and treat it when it reappears post transplant. Late antibody-mediated rejection can now be diagnosed and potentially treated in some grafts with deteriorating function. We now need trials to establish the effectiveness of the potential therapies to suppress alloantibody plasmapheresis, intravenous immunoglobulin, anti-CD20, and various maintenance immunosuppressives.
The emergence of criteria for diagnosing antibody-mediated rejection is a step toward the day when the clinician will diagnose T-cell- and antibody-mediated rejection in each organ by clinical, pathologic, and immunologic evidence; tailor treatment to the effector mechanism; and monitor the effect of intervention. This goal will require that we clarify the syndromes mediated by antibody and by effector T cells in each organ, and develop specific and sensitive tests for T-cell-mediated rejection. We should understand the mechanisms that determine whether effector T cells or antibody predominate: given that IgG alloantibody production depends on T-cell help through indirect recognition, the details of antigen presentation are of great interest. We should also define the role of endothelial protective genes and accommodation. But the many remaining tasks should not prevent us from appreciating the recent progress in defining the roles of alloantibody in clinical transplantation, and the landmark contributions of pioneers such as Gorer, Dausset, Van Rood and Terasaki.
- 3The detection of antigenic differences in mouse erythrocytes by the employment of immune sera. Br J Exp Pathol 1936; 17: 42–50..
- 4The cytotoxic activity of isoantibodies in mice. Transplant Bull 1956; 3: 142–143., .
- 8Patterns of permissible donor-recipient tissue transfer in relation to ABO blood groups. In: Experience in Renal Transplantation. Philadelphia: WB Saunders: 1964..